2.2 MetaDB

This program is designed to reformat all the tables of the database into a set of XML files which will be used by the analysis programs.

2.2.1 Command Line Options

Option	Usage	Type	Explanation
-e,--exp	required	string	Experiment file
-o,--output	required	string	Output XML directory
-m,--mapdb	required	string	The GLMdb directory
-q,--qtldb	optional	string	The QTLdb directory
-t,--tofile	optional	string	Trait Ontology file
-d,--mrkdico	optional	string	Marker dictionary file
--mrkup	optional	string	A file with the marker name to update
--mrkrm	optional	string	A file with the marker to remove
--chrm	optional	string	A file with the chromosome to remove

For example, suppose we have created a database which consists in the files exp.txt (experiment table), onto.txt (trait ontology), dico.txt (marker dictionary), and the directories glmdb/ and qtldb/ which contain the marker map and the QTL map tables relative to the mapping experiments specified in exp.txt. We assume that all the files and directories are included into the same directory, the database directory. Then we can create the XML repository, xmldb/, by invoking the command (in the database directory)

>java org.metaqtl.main.MetaDB \ -e exp.txt -t onto.txt -d dico.txt -m glmdb -q qtldb -o xmldb

This creates as many XML files in the xmldb/ directory as the number of mapping experiments. For a given mapping experiment all the information (parameters, markers, QTL, ...) are merged into a single XML file which stem name is the same than the mapping experiment name. The output directory xmldb/ also contains a file named ontology.xml which represents the trait ontology table (if this last one is not valid, the file is not created). Finally, using the marker dictionary, MetaDB converts all the marker names which match a synonym in the marker dictionary to their standard name. If you want to export these XML files into another file format, See Xml2A.

For the option --mrkup, the input file format must be a table with 4 columns seperated by a tabulation as follows:

     Exp1	Chrom1	mrk3	mrk3a
     Exp1	Chrom2	mrk4	mrk4b
     Exp2	Chrom1	mrk1	mrk1c
     ...

where the first column gives the name of the mapping experiment (must be the same than in the experiment table), the second the name of the chromosome, the third one the current name of the marker to update and the last one the new name of the marker.

For the option --mrkrm, the input file is the same than the previous one except that it contains only the first three columns.

Finally, for the option --chrm, the input file is the same than the previous one except that it contains only the first two columns.

Note that these three last options have no effect on the raw data file: the rules they define are only applied to the ouput XML data base.